Nakanos later modality allows types to express that the output of a function does not immediately depend on its input, and thus that computing its fixpoint is safe. This idea, guarded recursion, has proved useful in various contexts, from functional programming with infinite data structures to formulations of step-indexing internal to type theory. Categorical models have revealed that the later modality corresponds in essence to a simple reindexing of the discrete time scale. Unfortunately, existing guarded type theories suffer from significant limitations for programming purposes. These limitations stem from the fact that the later modality is not expressive enough to capture precise input-output dependencies of functions. As a consequence, guarded type theories reject many productive definitions. Combining insights from guarded type theories and synchronous programming languages, we propose a new modality for guarded recursion. This modality can apply any well-behaved reindexing of the time scale to a type. We call such reindexings time warps. Several modalities from the literature, including later, correspond to fixed time warps, and thus arise as special cases of ours.
تحميل البحث